Research web portal prototype

Interactive sequence-to-risk workbench for condensate and aggregation hypotheses.

Paste a protein sequence, choose an environment, and inspect protein-level scores, residue-level tracks, detected regions, and downloadable reports in one browser workflow.

0Residues

EarthEnvironment

--Confidence

Input

Protein sequence

0 residues

Example protein

Protein name UniProt ID

Sequence or FASTA

Environment

Transport model

default

Ready for live prediction

Select a demo protein or paste a sequence, then run prediction to generate an interactive dashboard.

Comparison

Earth vs selected environment

MSI --

Explainability

Feature importance

normalized

Residues

Score tracks

0 residues

Aggregation LLPS Disorder Low complexity

Detected intervals

Regions crossing thresholds

0 regions

Type	Start	End	Length	Mean	Max	Label

Structure

3D result viewer

score model

Run a prediction to view an input-sequence backbone with highlighted risk regions. Optionally load a PDB ID for an experimental structure.

FASTA alone does not contain a true folded structure. The input model is a coarse browser-generated backbone for residue-order visualization. Use a UniProt ID for AlphaFold or a PDB ID for an experimental structure; residue highlighting assumes numbering matches the submitted sequence.

Export

Download reports

JSON / CSV / HTML

SpaceCondenseNet